Miller Tanner saves time by structuring data

Trimming 800 hours a year with automated AI workflows

hero miller tanner image
miller tanner logo image
INDUSTRY

Professional services

COMPANY SIZE

100+ employees, all on Box

SOLUTIONS

Secure collaboration
Document management
Content workflows

PRODUCTS & INTEGRATIONS

Box AI API
Box Governance
Box Notes
Box Shuttle
Box Sign

800

hours saved yearly on manual entry

95%

of all content stored in Box

82K 

links shared

CHALLENGE
4 icon problem 2
  • Incoming PDFs from external travel agencies lacked a uniform format, making it tedious to extract information
  • The legacy ECM model included three different file systems, so content was siloed and hard to secure
  • Legacy ECM also made it challenging to innovate with new tools like AI
OUTCOME
4 icon solution 0
  • A custom workflow built on the Box AI API automatically structures unstructured information
  • Centralizing all files with Intelligent Content Management bolstered security and compliance
  • Box enables Miller Tanner to push the envelope of innovation

How to save hundreds of hours per year

If you’ve ever booked corporate travel, you’ve probably received a confirmation PDF by email. It’s a pretty simple document containing your itinerary, hotel booking, record locators, and other important details.

Simple for you, that is. But if you were a computer system handling volumes of PDFs for many people over time, you’d be stymied. That’s because every travel agency formats their booking PDF differently. To a computer, each travel itinerary is unstructured, inscrutable data, leaving the interpretation up to humans with a discerning eye.

That’s the quandary Miller Tanner Associates was facing. Since 1997, the full-service corporate event planning company has organized exceptional experiences for customers, including face-to-face, virtual, and hybrid meetings and events. An important component of this business is managing travel for up to 12K people attending meetings every year.

To do this, Miller Tanner works with multiple travel agencies globally — some long-term partners and some event one-offs. Certain attendees even book their own travel. Agents at Miller Tanner have to catalog the details of this travel in order to arrange ground transportation, hotel bookings, and other logistics. For a long time, human agents had to read every single incoming travel PDF and manually input the details into their proprietary operational database.

But now, by leaning into Intelligent Content Management from Box, Miller Tanner has been able to largely automate the transfer of data and save hundreds of hours a year.

 

Converting manual work to easy data extraction

In 2024, team members at Miller Tanner logged roughly 800 hours simply transcribing the information from incoming travel PDFs into their operational database. “That’s a lot of time, and it’s obviously a lot of expense,” says Daniel Conroy, Chief Digital Officer.

When Conroy and his team evaluated options, they figured there were three approaches the company could take:

  1. Keep going with manual entry, and keep adding more people to the team
  2. Insist on tabular entries, which would require every travel agency to submit a standardized format like CSV or spreadsheet 
  3. Take advantage of OCR — great when you have standardized forms and know exactly the format data will take 

 

Unfortunately, the last two workarounds wouldn’t work, because Miller Tanner didn’t have control over how travel agents submitted their information. As for the first, as Conroy puts it, “Manual entry is resource intensive, tedious as all getout, and not very scalable.” At the heart of the challenge was finding a way to structure the unstructured data coming in from all kinds of travel PDFs. That way was made possible with Intelligent Content Management, which enabled Conroy’s team to build a custom solution, using the Box AI API, that automatically extracts all of the unstructured data from incoming PDFs and catalogs it into the company’s operational database. 

 

Converting manual work to easy data extraction 

Automated extraction and API calls

As PDFs come in, a staff member uploads them to a folder on Box, and the automated extraction process kicks off. Miller Tanner has put two important API calls in place to make this happen:

  1. The Ask API provides some important info about the PDF — for instance, what type of document it is, which language it’s in, and which airports are listed (automatically converting names to airport codes) 
  2. The Extract API prompts for traveler name and details, agency information, flight information, train information, and other critical details 

 

The workflow then automatically saves the right data to staging tables, which can then be reviewed for accuracy. This system of automated extraction minimizes manual effort and ensures accuracy. Once a reviewer hits “Save,” the data automatically moves to the operational database.

Using Box to automate data extraction saves a ton of time and has a few other key benefits:

  • The ability to control the data and AI processing to ensure the security of personal data across different global regions 
  • Teamwork with an established partner with a legacy of trust, as opposed to partnering with an untested startup 
  • The ability to combine the Ask and Extract APIs to get the specific functionality they were looking for 

 

“We were really excited about it,” says Conroy, “because Box allowed us to extract a bunch of unstructured data and turn it into structured data without worrying about fixed layouts. And we gained the ability to interpret non-English language and characters. We have no control over what’s coming in, but we still need to be able to work with it.”

Using Box to automate data extraction from incoming PDFs gives Miller Tanner all the time back they used to spend on manually typing in this information, so they can serve customers much faster and more accurately. 

 

2.5+ terabytes of content moved to Box

This highly specific use of Intelligent Content Management is just one of the ways Miller Tanner uses Box to meet business goals. A global company with customers all over the world and a workforce that’s 100% remote, they also use Box for all kinds of content collaboration among team members, external partners, and customers.

Box has been a dramatic improvement over the previous external file-sharing tool. As Conroy says, “I felt like we were just chasing our tail, frankly, trying to find the latest version. Collaboration was difficult because we weren’t connected well. We were spending a lot of time and money just trying to maintain the system.”

With the old file-sharing paradigm, Miller Tanner had three different file systems. With Box, they have one. Using Box Shuttle, they were able to successfully migrate about two-and-a-half terabytes of content into Box, and that volume has grown seven- or eightfold since. But because people now have a central, compliant place they can put the content they gather and create, Miller Tanner’s content volume has grown securely, in compliance, and without fragmenting content all over the place.

As a result, Conroy says, “Usage has shot up, and now we know where everything is. Box is where everything goes, at the end of the day. It’s part of the way we work.”


Zero compromise on security

Because travel data is inherently personally identifying data, Miller Tanner has to prioritize data security and compliance while working with an AI solution (and any technology solution). As Conroy says, “I wanted people to be able to get their jobs done, without creating extra work being worried about security.”

Box gave Miller Tanner secure sharing capability, granular permissioning, versioning and auditing, and more. Box was also extensible to other applications in their technology repertoire, so that employees working in tools such as Word or Excel could still count on content being safely housed on Box.

One of the biggest things Conroy appreciates about Box is the conservative approach to security even when experimenting with AI: “We can’t just turn this thing on and vacuum up all this information. We honor our organizational controls.”

 

Zero compromise on security 

The payoff of Intelligent Content Management

Miller Tanner had myriad reasons to switch to Box, but the real payoff has come with the custom AI capability. Because Box is inherently an AI-powered platform, Miller Tanner was at an advantage over competitors who weren’t set up to use AI with their existing tech frameworks. They were able to work with their Box account rep to help build out a customized solution. “This unlocked everything,” says Conroy. “Now, we were able to leverage this wonderful platform we use for everything — and bring a level of intelligence that we just didn’t have. And stay within Box to do it.”

Conroy looks forward to developing further automated workflows with Intelligent Content Management. That might include workflows that are part of the employee onboarding process, for instance, and integrating Salesforce with content on Box via the Box for Salesforce integration.

Miller Tanner also plans to tap into the power of data extraction to help find information across files. Conroy says, “I’d like to turn this technology toward other unstructured data sources such as information exchanges with hotel partners.”

The top three things Miller Tanner loves about Box, according to Conroy:

  1. Ease of customization: “They’ve done a very good job of making it easy for us to work with the data we have in the systems that we have.” 
  2. Reliability: “Something that Box has done really well. We haven’t had any major issues, and that reliability is huge for our business.” 
  3. Innovation: “Box is pushing the envelope in a meaningful way that makes it easier for all of us to access the content that’s in there.” 

 

 

Box brought a level of intelligence that we just didn’t have.

— Daniel Conroy, Chief Digital Officer, Miller Tanner

 

See what Box can do for you

Explore how other customers are using Box

travers smith card
Travers Smith

Saving up to $5K per legal case by applying a highly secure data model to a 200+ year-old firm

Read story
Vornado card
Vornado

Building on Class A content to construct real estate solutions with Intelligent Content Management

Read story
finn partners card
Finn Partners

Scaling to dozens of offices with content consistency so acquisition goes smoothly time after time

Read story